Adobe Flash SEO Critique

While I commend Adobe for its recent efforts to help engines index textual content locked in Flash, I have issues with the new "SEO Technology Center."

For example, the following video by one of Adobe's Senior Technology Evangelists states that "...rises to the top of the heap in the Google..." for [Duanes World] thanks to Adobe's new headless Flash player technology "Ichabod." According to Adobe's Evangelist, "Duane" could only be visible to Googlebot by having Ichabod change states in the Flash file, therefore exposing "Duane" as textual content. Unfortunately this is not correct as the cached version of the page from Google's SERP states "These terms only appear in links pointing to this page".

As shown in the Google SERP, "Duane" appears to Googlebot only in links pointing at and not in the Flash file as the video claims. Using the advanced "site:" operator to search for [Duane] within shows a number of pages with links pointing to AdobeTV using "Duane" as anchor text. Because these links use #anchors (fragment identifiers) which Googlebot ignores, in URLs, Google "credits" keyword relevancy to the root instead of the intended target URL.

As a result of this misallocation, ranks for [Duanes World]. Just as keyword relevancy is being misallocated in this case, so is PageRank as illustrated by the URL PR9 when compared with the Flash PR8. In fact, the SERP TITLE and snippet aren't from AdobeTV but rather DMOZ. To see this compare "Adobe TV" as seen in the SERP with "AdobeTV" seen on the page by users with Flash and JavaScript enabled. This is due to AdobeTV's use of dynamic JavaScript TITLE elements in (X)HTML.

With on page factors out of the way, as a side note it's also worth mentioning that the search engine results page in Adobe's video is based on the user's prior search history, while logged into a Google Account and Searchwiki appears to be activated. These personalization settings can all act to throw off the data in such an experiment.

This post isn't intended to bash Adobe but rather to point out some critical errors in their research. Please don't get me wrong, I'm a huge fan of Adobe and have been for years. I think they make great products and appreciate all of the hard work done by Adobe's team of Evangelists. I understand that Adobe Evangelists are experts at Flash but, when it comes to SEO for Flash and interpreting Google SERPs, wish people wouldn't take their opinions blindly as being fact.

For the handful of us with expertise in SEO for Flash, it's a little awkward having to tell clients that Adobe's information isn't entirely correct. Either way, it would be nice to see more research as well as accurate and up-to-date information in Adobe's SEO Technology Center. It would also be great to see some of these best practices implemented at

3 thoughts on “Adobe Flash SEO Critique

  1. technoracle


    While I may have said “could only be visible to Googlebot by having Ichabod change states in the Flash file, therefore exposing “Duane” as textual content” in the video (I haven’t checked it), there are also other reasons why it may come up. In this case though, we do know that the states were changed and I have run some other tests.

    The most obvious reason today is that link equity is helping Google find the site relevant. This includes that fact that points at it in addition to many other sites. The reasons you cite are also very likely a much larger impact on the search results now. Another reason is that people searching for “duanes world” also regularly chose to search for “adobe + duanes world” and also click through on the links in the results pages.

    I generally agree with your advice though. Too many people spend time and effort trying to optimize Flash for search engines in the wrong places. The Adobe TV site has many errors on it which are picked up by Google too. In fact, if you tried a search on for “duane”, our own search feature used to not return any results for that search despite the fact “duane” is in text in some of the assets.

    Nevertheless, I am glad to have met (virtually) someone else who understands this space as well as you. I am running some advanced tests for determining how well google follows links and how far it reaches into a swf. Additionally, the part that is not yet known to many of us is how Google will actually initially rank from swf content vs the same content from HTML. My tests have unique keywords in both HTML and swf in the same font/size context in two different URL’s/IP addresses. The intent is to see how the ranking works for . I obviously cannot share this as it has to be uncontaminated by people searching and clicking or linking to it. I am curious to understand what weighting swf gets compared to text/HTML.

    Another thing I am trying to understand is how far Google will reach into swf’s. Usually about 200 unique words are the limit from most PR <4 sites in Google (of course throwing out stopwords). Initial research has shown that swf might do less but it is unclear because the environment in often contaminated by link equity and click traffic (hence proving that these two mechanisms are the best chance for SEO). Not sure how to fully test this but I am trying to get a range of pages with swf content to get ranked at various level of pagerank by Google then seed the content with terms to see what the threshold is.

    BTW – love your blog.


  2. beussery

    Post author

    Thought I was pretty clear in my post but, this isn’t the type of “link equity” to be promoting and there is no evidence to support the claims made in the video.

Comments are closed.